Michael G. Schwern [Tue, 10 May 2005 18:10:47 +0000]
Hints changes for OS X 10.4
Message-ID: <
20050511011047.GA23955@windhund.schwern.org>
p4raw-id: //depot/perl@24446
Rafael Garcia-Suarez [Wed, 11 May 2005 07:54:19 +0000]
Include vim/emacs modelines in generated files to open them
in read-only mode. Make vi modelines compatible with non-vim
vi versions.
p4raw-id: //depot/perl@24445
Jarkko Hietaniemi [Tue, 10 May 2005 19:44:00 +0000]
potential [PATCH] Tru64 crank up strictness
Message-Id: <
4ADE5AAD-27CB-4F9E-BEC7-41DAA7671108@iki.fi>
p4raw-id: //depot/perl@24444
Rafael Garcia-Suarez [Tue, 10 May 2005 20:27:16 +0000]
"make regen" should run reentr.pl
p4raw-id: //depot/perl@24443
Rafael Garcia-Suarez [Tue, 10 May 2005 20:08:27 +0000]
Regenerate reentr.[ch]
p4raw-id: //depot/perl@24442
Steve Peters [Tue, 10 May 2005 12:44:13 +0000]
Re: [PATCH] reentr.h changes so threaded Perl's compile on OpenBSD 3.7
Message-Id: <
20050510174413.GA19686@mccoy.peters.homeunix.org>
p4raw-id: //depot/perl@24441
Rafael Garcia-Suarez [Tue, 10 May 2005 16:51:02 +0000]
Add editor boilerplates to all C files
(except the generated ones)
p4raw-id: //depot/perl@24440
Nicholas Clark [Tue, 10 May 2005 15:38:07 +0000]
s/sv_setpv(sv,"")/sv_setpvn(sv,"",0)/
plus a couple of 1 byte sv_setpv()s too.
p4raw-id: //depot/perl@24439
Nicholas Clark [Tue, 10 May 2005 14:47:58 +0000]
As we have the length of the string easily to hand, no reason not to
use it.
p4raw-id: //depot/perl@24438
Steve Peters [Tue, 10 May 2005 09:41:01 +0000]
perlop.pod nit
Message-Id: <
20050510144101.GA21362@mccoy.peters.homeunix.org>
p4raw-id: //depot/perl@24437
Nicholas Clark [Tue, 10 May 2005 11:03:49 +0000]
Experiment with using the ptr_table code to hold the seen hash
p4raw-id: //depot/perl@24436
Nicholas Clark [Tue, 10 May 2005 09:17:32 +0000]
Simplify the debug and assert logic.
p4raw-id: //depot/perl@24435
Nicholas Clark [Tue, 10 May 2005 09:03:11 +0000]
Change 24165 missed one s/class/classname/ in the debugging code.
p4raw-id: //depot/perl@24434
Steve Peters [Mon, 9 May 2005 21:10:49 +0000]
reentr.h changes so threaded Perl's compile on OpenBSD 3.7
Message-Id: <
20050510021049.GA20147@mccoy.peters.homeunix.org>
p4raw-id: //depot/perl@24433
Jan Dubois [Mon, 9 May 2005 19:29:06 +0000]
Avoid USE_SITECUSTOMIZE redefined warnings
Message-Id: <
200505100229.j4A2TGm1019135@smtp3.ActiveState.com>
p4raw-id: //depot/perl@24432
Andy Lester [Mon, 9 May 2005 17:45:00 +0000]
attributes-two.patch, the next batch
Message-ID: <
20050509154518.GA18273@petdance.com>
p4raw-id: //depot/perl@24431
Rafael Garcia-Suarez [Mon, 9 May 2005 19:19:20 +0000]
Fix typo (bug #35368)
p4raw-id: //depot/perl@24430
Nicholas Clark [Mon, 9 May 2005 16:33:57 +0000]
Update the comment describing arenas.
p4raw-id: //depot/perl@24429
Nicholas Clark [Mon, 9 May 2005 11:22:19 +0000]
Hoist the static "AUTOLOAD" into a static const variable. Makes the
object file slightly smaller, and also increases the proportion of
read only data section.
p4raw-id: //depot/perl@24427
Rafael Garcia-Suarez [Mon, 9 May 2005 11:11:35 +0000]
Upgrade to Unicode::Collate 0.50
p4raw-id: //depot/perl@24426
Nicholas Clark [Mon, 9 May 2005 10:10:42 +0000]
Change 24413 should have updated makedef.pl with the knowledge that 2
symbols are ithreads only.
p4raw-id: //depot/perl@24425
Nicholas Clark [Mon, 9 May 2005 10:00:37 +0000]
Change 24420 should have hand updated globvar.sym to reflect the
renaming of the global variable.
p4raw-id: //depot/perl@24424
Nicholas Clark [Mon, 9 May 2005 08:48:46 +0000]
Simplify S_hv_notallowed slightly by passing a prebuilt message
template. (Also slightly smaller object code)
p4raw-id: //depot/perl@24423
Steve Peters [Sun, 8 May 2005 20:49:17 +0000]
Sv_*set() doc's and extra const's for the SvPVX_const() tasks
Message-ID: <
20050509014917.GA24132@mccoy.peters.homeunix.org>
p4raw-id: //depot/perl@24422
Rafael Garcia-Suarez [Mon, 9 May 2005 07:29:33 +0000]
Revert to Andy's first solution for nonnull attributes;
add the missing defines for non-threaded perls in perl.h
p4raw-id: //depot/perl@24421
Nicholas Clark [Sun, 8 May 2005 22:09:18 +0000]
Perl_croak uses perl's printf, so can pass in SVs direct - no need for
SvPV and a %s format.
p4raw-id: //depot/perl@24420
Nicholas Clark [Sun, 8 May 2005 21:40:29 +0000]
There's no need to set a reference count for the new SV heads, as it's
always set to 1 at uprooting time.
But set it to zero when DEBUGGING to make things clearer. Plus fix a
bug introduced by change 22945, where the last SV head had an
uninitialised reference count.
p4raw-id: //depot/perl@24419
Nicholas Clark [Sun, 8 May 2005 19:45:30 +0000]
The idea is that when you find something, you stop looking.
( http://use.perl.org/comments.pl?sid=26369&cid=40183 )
p4raw-id: //depot/perl@24418
Dave Mitchell [Sun, 8 May 2005 14:39:19 +0000]
t/TEST's new error messages now include the prefix 'FAILURE--'
p4raw-id: //depot/perl@24417
Rafael Garcia-Suarez [Sun, 8 May 2005 12:19:51 +0000]
Tweak the insertion of __attribute__((nonnull())) declarations
in proto.h so it compiles with and without threads with gcc 3.4
p4raw-id: //depot/perl@24416
Andy Lester [Wed, 4 May 2005 23:55:00 +0000]
GCC attributes!
Message-ID: <
20050504215540.GA20413@petdance.com>
p4raw-id: //depot/perl@24414
Nicholas Clark [Sat, 7 May 2005 16:08:16 +0000]
The ptr_table arena variables and code is only needed for ithreads.
p4raw-id: //depot/perl@24413
Dave Mitchell [Sat, 7 May 2005 12:57:06 +0000]
while (my $x ...) { ...; redo } shouldn't undef $x.
In the presence of 'my' in the conditional of a while(), until(),
or for(;;) loop, add an extra scope to the body so that redo
doesn't undef the lexical
p4raw-id: //depot/perl@24412
Dave Mitchell [Sat, 7 May 2005 11:06:16 +0000]
add 1.875c to the list of supported bisons
p4raw-id: //depot/perl@24411
Nicholas Clark [Fri, 6 May 2005 22:17:04 +0000]
Get the correct refcount on the temporary assignment to PL_sigwarn so
that other code assigning to $SIG{__WARN__} doesn't cause a premature
free.
p4raw-id: //depot/perl@24410
Nicholas Clark [Fri, 6 May 2005 19:08:35 +0000]
Ooops. Wise man tests with ithreads before commiting. D'oh!
p4raw-id: //depot/perl@24409
Nicholas Clark [Fri, 6 May 2005 18:38:45 +0000]
Change types of PL_he_arenaroot and PL_pte_arenaroot to avoid casting.
p4raw-id: //depot/perl@24408
Rafael Garcia-Suarez [Fri, 6 May 2005 15:37:30 +0000]
Upgrade to PathTools 3.07
p4raw-id: //depot/perl@24407
Abhijit Menon-Sen [Fri, 6 May 2005 15:17:26 +0000]
Don't pretend SIGZERO exists [#35291].
p4raw-id: //depot/perl@24406
Nicholas Clark [Fri, 6 May 2005 13:11:54 +0000]
Make the arena size changeable at compile time, and up the default by
a factor of 4.
p4raw-id: //depot/perl@24405
Nicholas Clark [Fri, 6 May 2005 12:34:36 +0000]
Allocate pointer table entries (for ithread cloning) from an arena
p4raw-id: //depot/perl@24404
Dave Mitchell [Thu, 5 May 2005 20:52:11 +0000]
avoid accessing a just-freed SV (keep ponie happy)
p4raw-id: //depot/perl@24402
Rafael Garcia-Suarez [Thu, 5 May 2005 17:12:38 +0000]
Missing files in Test::Simple
p4raw-id: //depot/perl@24401
Nicholas Clark [Thu, 5 May 2005 15:17:23 +0000]
Bump versions of non dual-life modules
p4raw-id: //depot/perl@24399
Nicholas Clark [Thu, 5 May 2005 14:54:20 +0000]
Bump version numbers
p4raw-id: //depot/perl@24398
Rafael Garcia-Suarez [Thu, 5 May 2005 14:41:55 +0000]
Upgrade to Test::Simple 0.60
p4raw-id: //depot/perl@24397
Nicholas Clark [Thu, 5 May 2005 14:28:42 +0000]
Bump B version numbers
p4raw-id: //depot/perl@24396
Dave Mitchell [Thu, 5 May 2005 12:06:32 +0000]
stop Math/BigInt/t/bare_mbf.t producing spurious test output
p4raw-id: //depot/perl@24393
Dave Mitchell [Thu, 5 May 2005 12:01:51 +0000]
strictifying t/TEST stopped it handling '1..n todo' correctly
(made scope of %toto too narrow)
p4raw-id: //depot/perl@24392
Dave Mitchell [Thu, 5 May 2005 00:34:50 +0000]
make t/TEST detect various failure modes in testfile outputs
in particular:
multiple leader lines
leader line not first or last line of file
unrecognised lines
out-of-sequence test numbers
p4raw-id: //depot/perl@24390
Dave Mitchell [Wed, 4 May 2005 22:14:54 +0000]
strictify t/TEST
make most variables lexical, narrow their scope, and give some of
them more meaningful names. Also fix a harmless bug where the wrong
filehandles are closed (A+B rather than U+A)
p4raw-id: //depot/perl@24389
Rafael Garcia-Suarez [Wed, 4 May 2005 20:50:59 +0000]
Document the unportability of atan2() edge cases
and remove unportable tests (by Steve Peters)
p4raw-id: //depot/perl@24388
Dave Mitchell [Wed, 4 May 2005 20:30:00 +0000]
add test for [perl #34682] leaving eval via last in inner runops
p4raw-id: //depot/perl@24387
Rafael Garcia-Suarez [Wed, 4 May 2005 18:31:19 +0000]
Fix test numerotation
p4raw-id: //depot/perl@24386
Dave Mitchell [Wed, 4 May 2005 15:52:11 +0000]
make goto.t use test.pl, strict and warnings
p4raw-id: //depot/perl@24385
Dave Mitchell [Wed, 4 May 2005 14:01:40 +0000]
[perl #35214] SEGV when next is followed by a goto
next and redo didn't restore PL_curcop
p4raw-id: //depot/perl@24384
Offer Kaye [Wed, 4 May 2005 15:10:37 +0000]
perlvar.pod verbatim paragraph first line does not start with a space
Message-Id: <
56942505050405101bfe678d@mail.gmail.com>
p4raw-id: //depot/perl@24383
Rafael Garcia-Suarez [Wed, 4 May 2005 11:21:54 +0000]
Sync synopses
p4raw-id: //depot/perl@24382
Dave Mitchell [Tue, 3 May 2005 22:10:45 +0000]
document the internals of exception handling
p4raw-id: //depot/perl@24381
Rafael Garcia-Suarez [Tue, 3 May 2005 15:43:34 +0000]
Fix getting prototype of builtins mkdir and setpgrp.
Hardcode special exceptions for exec and system, instead
of trying to be clever and failing.
p4raw-id: //depot/perl@24380
Rafael Garcia-Suarez [Tue, 3 May 2005 12:18:50 +0000]
Error in the latest FindBin patch, noticed by Nicholas
p4raw-id: //depot/perl@24379
Rafael Garcia-Suarez [Tue, 3 May 2005 12:15:45 +0000]
mkdir without arguments now defaults to $_
p4raw-id: //depot/perl@24378
Nicholas Clark [Tue, 3 May 2005 12:00:25 +0000]
ok($a == $b) better written as cmp_ok($a, '==', $b)
(The latter gives more diagnostics in the case of failure)
p4raw-id: //depot/perl@24377
Rafael Garcia-Suarez [Tue, 3 May 2005 09:45:24 +0000]
Convert to test.pl
p4raw-id: //depot/perl@24376
Rafael Garcia-Suarez [Tue, 3 May 2005 08:53:25 +0000]
Fix for [perl #34252] Access rights in FindBin::Bin
At least on my platform, Cwd::getcwd doesn't find the current
directory if it has no access to it. Try harder with Cwd::cwd.
p4raw-id: //depot/perl@24375
Steve Hay [Tue, 3 May 2005 08:51:24 +0000]
Fix croak() and confess() so that they don't clobber $!
(plus tests to check this)
p4raw-id: //depot/perl@24374
Nicholas Clark [Tue, 3 May 2005 08:45:28 +0000]
PL_mess_sv is always >= SVt_PVMG, so no need for an if() test
p4raw-id: //depot/perl@24373
Sergey Skvortsov [Mon, 17 Jan 2005 09:02:33 +0000]
[perl #33809] optimize macro dXSARGS
From: Sergey Skvortsov (via RT) <perlbug-followup@perl.org>
Message-ID: <rt-3.0.11-33809-105714.16.
4209917806492@perl.org>
p4raw-id: //depot/perl@24372
Steve Peters [Fri, 15 Apr 2005 09:10:54 +0000]
IEEE math for the masses
Message-ID: <
20050415141054.GA12749@mccoy.peters.homeunix.org>
(tests added to t/op/exp.t)
p4raw-id: //depot/perl@24371
Rafael Garcia-Suarez [Tue, 3 May 2005 06:30:47 +0000]
Convert to test.pl
p4raw-id: //depot/perl@24370
Nicholas Clark [Mon, 2 May 2005 20:45:00 +0000]
A slightly better layout for the diagram.
p4raw-id: //depot/perl@24369
Rafael Garcia-Suarez [Mon, 2 May 2005 20:10:34 +0000]
Clarify the definition of the 'w' pack format, as suggested by
Alexey Toptygin.
p4raw-id: //depot/perl@24368
Rafael Garcia-Suarez [Mon, 2 May 2005 17:10:19 +0000]
For perls where pids and ppids are cached, when the ppid of
the perl process becomes 1, refresh the ppid cache (this may
indicate that the parent process has died.)
p4raw-id: //depot/perl@24367
Nicholas Clark [Wed, 27 Apr 2005 14:57:51 +0000]
Change inheritance hierarchy of B subclassses :
Subject: @B::NV::ISA = 'B::IV';
Message-ID: <
20050427135750.GD94420@plum.flirble.org>
p4raw-id: //depot/perl@24366
Andy Lester [Sun, 1 May 2005 15:07:55 +0000]
printf.patch: Adding printf checking for gcc
Message-ID: <
20050501200755.GA30259@petdance.com>
p4raw-id: //depot/perl@24365
Rafael Garcia-Suarez [Wed, 3 Nov 2004 16:38:01 +0000]
Re: [perl #31793] Data::Dumper: Useqq interacts badly with overloading
Message-ID: <
20041103163801.6839be30@valis.local>
p4raw-id: //depot/perl@24364
Dave Mitchell [Mon, 2 May 2005 14:27:20 +0000]
Better fix for #8738 (Core dump in 'leavetry')
When in an inner runops loop (eg via a tie or sort), an eval
needs a new JMPENV pushing by S_docatch. If an exception is raised,
control is returned to S_docatch, and it must determine whether
the eval that trapped the exception is an inner eval or an outer
one. In the former case, restart the loop, in the latter case,
rethrow the exception. This is determined by whether we are still
at the same PL_curstackinfo level. This fails in the case of
SPLICE(), which pushes a new SETJMP and runops level, but not a
new stackinfo level. There may be other code which does similar.
The solution is to store the current value of PL_top_env in each
pushed CxEVAL, and see if it's still the same as PL_top_env when
the exception is handled.
p4raw-id: //depot/perl@24363
Dave Mitchell [Mon, 2 May 2005 13:51:42 +0000]
Revert change #15705 (Core dump in 'leavetry')
This fixes bug #34682, reintroduces bug #8738 (ID
20020301.011),
and reintroduces an eval optimisation for innter runops levels
p4raw-link: @15705 on //depot/perl:
8bffa5f8f4dd0cd203052722c9fcfd899f51d033
p4raw-id: //depot/perl@24362
Dave Mitchell [Mon, 2 May 2005 13:29:17 +0000]
move the SETJMP exception-handing definitions from scope.h to cop.h
so that a JMPENV* entry can be added to struct block_eval
p4raw-id: //depot/perl@24361
Rafael Garcia-Suarez [Mon, 2 May 2005 09:37:04 +0000]
Fix [perl #35162] $SIG{__DIE__} = 'IGNORE' is base.pm is illegal
p4raw-id: //depot/perl@24360
Craig A. Berry [Fri, 29 Apr 2005 17:22:48 +0000]
undo pod utilities regression in 24034
From: "Craig A. Berry" <craigberry@mac.com>
Message-ID: <
4272B3B8.6030603@mac.com>
p4raw-id: //depot/perl@24359
Steven Philip Schubiger [Mon, 2 May 2005 01:14:29 +0000]
uninitialized warnings in regcomp
Message-Id: <
200505012314.j41NETLk018737@accognoscere.homeunix.org>
p4raw-id: //depot/perl@24358
Nicholas Clark [Sun, 1 May 2005 22:16:16 +0000]
Explode if anyone attempts to sv_upgrade PL_mess_sv.
Should this be a panic: ?
p4raw-id: //depot/perl@24357
H.Merijn Brand [Fri, 29 Apr 2005 15:32:17 +0000]
Put back INSTALL_PREFIX and INSTALL_PREFIX_EXP as they were
unintendedly removed by patch #23435
p4raw-id: //depot/perl@24355
Sébastien Aperghis-Tramoni [Thu, 28 Apr 2005 02:24:04 +0000]
Small patch to perlport.pod
Message-Id: <
D38E89F3-B77B-11D9-B91D-000502F3279F@free.fr>
p4raw-id: //depot/perl@24354
Jarkko Hietaniemi [Fri, 29 Apr 2005 11:02:01 +0000]
one cannot "static" exported functions
Message-ID: <
B356D8F434D20B40A8CEDAEC305A1F2453D705@esebe105.NOE.Nokia.com>
p4raw-id: //depot/perl@24353
Steve Hay [Fri, 29 Apr 2005 08:24:50 +0000]
Skip PL_sig_sv when PERL_IMPLICIT_CONTEXT is defined
(rather than when PERL_IMPLICIT_SYS is defined)
p4raw-id: //depot/perl@24352
Nicholas Clark [Thu, 28 Apr 2005 22:28:53 +0000]
Refactor Perl_sv_utf8_upgrade_flags to use SvPV_free
p4raw-id: //depot/perl@24351
Jarkko Hietaniemi [Wed, 27 Apr 2005 17:35:27 +0000]
(in blead) a missing const in a prototype
Message-ID: <
B356D8F434D20B40A8CEDAEC305A1F2453D6ED@esebe105.NOE.Nokia.com>
p4raw-id: //depot/perl@24350
Steve Hay [Thu, 28 Apr 2005 16:38:51 +0000]
Silence a compilation warning on Win32
p4raw-id: //depot/perl@24349
Nicholas Clark [Thu, 28 Apr 2005 11:22:15 +0000]
Add a new macro SvPV_free() which undoes OOK and free()s the PVX(),
becase there's a lot of code around that calls SvOOK_off(), memmov()s
the buffer, then promptly free()s it. So avoid the needless memmov().
p4raw-id: //depot/perl@24348
Steve Hay [Thu, 28 Apr 2005 11:13:55 +0000]
Remove temporary clean-up from Win32 makefiles
Change 22501 temporarily added some clean-up to the Win32 makefiles
which was only meant to stay until cpan rt #5616 was fixed. It is
now fixed (in EU-MM 6.22, which in turn has been in blead for a
while now), so these hacks are no longer required.
This change is also good for maint, which currently contains EU-MM
6.17, which didn't create the blibdirs.exists files in question
anyway.
p4raw-id: //depot/perl@24347
Rafael Garcia-Suarez [Thu, 28 Apr 2005 09:29:07 +0000]
Remove confusing punctuation
(spotted by David Rigaudiere)
p4raw-id: //depot/perl@24345
Nicholas Clark [Thu, 28 Apr 2005 09:29:06 +0000]
Calling sv_backoff() on something that's about to be free()d will
memmov() memory that's about to be freed. Seems wasteful.
p4raw-id: //depot/perl@24344
Steve Hay [Thu, 28 Apr 2005 08:44:28 +0000]
Fix perlsio_binmode() export breakage caused by change #24338
Define and export perlsio_binmode() based on the same condition,
namely, that USE_PERLIO is defined and USE_SFIO is not.
p4raw-link: @24338 on //depot/perl:
de009b76d60bdeb88f1d812ac755ae225805a071
p4raw-id: //depot/perl@24341
Nicholas Clark [Thu, 28 Apr 2005 08:34:57 +0000]
The if() clause is the wrong way round, so OOK can never be true in
this case. Given that OOK is never undone, we could never have been
correctly upgrading a PV with OOK to HV or AV. I believe that only
SVt_NULL is the only type ever upgraded to HV or AV.
p4raw-id: //depot/perl@24340
Rafael Garcia-Suarez [Wed, 27 Apr 2005 20:24:30 +0000]
A couple of warning fixes by Gisle
p4raw-id: //depot/perl@24339
Andy Lester [Wed, 27 Apr 2005 10:02:43 +0000]
perlio-two.patch: More warnings squashed, more consts
Message-ID: <
20050427150243.GA21883@petdance.com>
p4raw-id: //depot/perl@24338
Steve Hay [Wed, 27 Apr 2005 08:35:40 +0000]
Win32 MM test fix
From: Michael G Schwern <schwern@pobox.com>
Date: Mon, 18 Apr 2005 14:30:19 -0700
Message-ID: <
20050418213019.GA15797@windhund.schwern.org>
Subject: Re: Win32 MM test fix
From: demerphq <demerphq@gmail.com>
Date: Tue, 19 Apr 2005 00:57:59 +0200
Message-ID: <
9b18b31105041815573e10dc10@mail.gmail.com>
p4raw-id: //depot/perl@24334
Andy Lester [Fri, 22 Apr 2005 17:39:47 +0000]
XS() __attribute__
Message-ID: <
20050422223947.GA4681@petdance.com>
p4raw-id: //depot/perl@24333
projects / p5sagit/p5-mst-13.2.git / log